1. Technical Field
The present invention relates to an image surveillance system and, more particularly, to an integrated image surveillance system and an image synthesis method thereof, wherein the system is capable of integrating video images captured by multiple cameras of an area into a large-coverage image of the area, as well as changing between different surveillance angles rapidly and smoothly, so as to facilitate monitoring of the area.
2. Description of Related Art
Recently, public security has become an issue due to rising crime rates. To ensure personal safety, people resort to a variety of security systems, among which intelligent video surveillance systems have drawn more and more attention. For a conventional large-scale video surveillance system to monitor an area effectively, several cameras are deployed at different locations in the monitored area so that images taken by the cameras cover the entire area. Hence, the larger the monitored area is, the more cameras need to be installed. The images captured by the cameras are displayed in multiple windows on a screen, as shown in FIG. 1, and monitored by a security guard. Since the views are separated and the lack of spatial relation information between the adjacent screens, the security guard may easily be distracted and miss important information and details in the images. The multiple screens may also confuse the security guard visually. Moreover, it is difficult to track an object of interest through the multiple screens and the security guard may become disoriented and misjudge in the course of tracking.